Data and the effective, timely movement of data has become the lifeblood of many modern enterprises. Unfortunately, network infrastructure limitations are impinging upon the ability of enterprises to timely move data. Enterprises that require data to be delivered within very strict time requirements are being most severely impacted by these network infrastructure limitations. To guarantee the timely delivery of their data, these enterprise often are forced to pay steep prices. Moreover, network providers are being forced continually to upgrade their infrastructure to supply even the basic services to their customers. Accordingly, the networking community is searching for a method and system to better and more economically utilize the existing network infrastructure, thereby improving the transfer of data and reducing the associated cost.
With regard to the actual transmission of data, enterprises are searching for a way to pay only for the bandwidth that they use. In essence, they are looking to optimize the use of bandwidth. Presently, an enterprise that requires the bandwidth provided, for example, by a T1 line may be forced to rent a dedicated T1 line for an entire month even though the enterprise may only need the T1 line for a few days within that month. Renting the T1 line for such an extended period is wasteful on two fronts. First, the enterprise is paying for many days of service that it does not use. Preferably, an enterprise should only pay for the service that it actually uses. Second, the bandwidth available on the T1 line for those unused days is wasted because other enterprises do not have access to it. If this unused bandwidth can be captured and made available to other enterprises, the existing network infrastructure can be better utilized to meet the demands of more enterprises. Unfortunately, no device or method exists to effectively optimize the provisioning of bandwidth. Thus, even though dedicated lines are expensive and cumbersome, companies requiring rapid, predictable transfer of data presently have no other acceptable option.
With regard to the actual routing of data, enterprises are searching for an efficient way to route data based upon priority. Several methods have been developed to aid in routing data based upon priority. These present methods, however, are not completely satisfactory. For example, a feature called weighted fair queuing can be enabled on modern routers. This feature requires that the router read a precedence bit from each packet being passed through the router and then queue lower priority packets while routing higher priority packets. Although weighted fair queuing can be effective, it requires a great deal of processing power and memory within each router, and when enabled for a long period of time, weighted fair queuing can cause a router to crash. Additionally, because weighted fair queuing requires a router to analyze each packet that it receives, it slows the overall operation of the router and, thus, the network.
Because router configuration is a somewhat complicated and time consuming process, especially if numerous routers are involved, network administrators tend to configure routers in the network either to use weighted fair queuing at all times or not to use weighted fair queuing at all. In reality, most network administrators would like to use weighted fair queuing some of the time and disable it the rest of the time. Unfortunately, selectively enabling and disabling weighted fair queuing is so cumbersome that it cannot be effectively implemented. Accordingly, a device and method are needed to selectively enable router optimization techniques such as weighted fair queuing.
Because the present network technology suffers from significant drawbacks, a solution is needed that can efficiently and effectively optimize a network to enable a more efficient transfer of data. In particular, a system and method are needed in which both the bandwidth usage and/or the router performance can be easily optimized. Such a system and method would not only address the needs of the network community, but also provide new advantages such as content transfer optimization.